Abstract
<title>Abstract</title> <p>Intermittent theta burst stimulation (iTBS) of the left dorsolateral prefrontal cortex (DLPFC) is an established depression treatment. However, antidepressant outcomes demonstrate significant variability among therapy plans and individuals. We still lack physiological readouts of when and how stimulation engages the targeted region and network on an individual level. Interleaved TMS-fMRI can map stimulation effects across the brain, but analyses still model the response from stimulation timing convolved with a canonical hemodynamic response function. This assumes that repeated TMS trains evoke stable BOLD responses within iTBS sessions and across individuals. Here, we tested whether this assumption misses relevant target engagement by directly measuring local hemodynamic responses during stimulation. We combined interleaved prefrontal iTBS-fMRI with simultaneous functional near-infrared spectroscopy (fNIRS), using fMRI for spatially resolved whole-brain mapping and fNIRS for independent tracking of direct hemoglobin signals at the stimulation site. In nine healthy participants receiving standard therapeutic iTBS, the conventional fMRI model did not reveal significant target activation at an individual level. Conversely, the model using fNIRS-informed individual oxygenated hemoglobin time courses from the stimulated cortex revealed significant responses in targeted left DLPFC and connected network nodes. The measured hemodynamic responses varied substantially within and across participants and deviated from canonical model assumptions. These findings suggest that an absent mean local BOLD response during TMS-fMRI does not necessarily indicate absent target engagement. Direct measurement of this variability improves fMRI sensitivity and provides a way to reveal network engagement during prefrontal iTBS to link acute target engagement to individualized neuromodulation strategies for depression.</p>